Takahiro Suzuki, Satomi Yamaya and Masaru Ishida
Chemical Resources Laboratory, Tokyo Institute of Technology,
Yokohama, Japan

The development of efficient removal processes for persistent chemicals in the environment including petroleum hydrocarbons is increasingly important in situ treatment for accidental release of petroleum and in wastewater treatment in the chemical industry.  Research has been conducted to remove organic compounds such as benzene, toluene, phenol and halogenated hydrocarbons using aerobic or anaerobic microorganisms in laboratory experiments [1].  So far, several types of bioreactor for the biodegradation of hydrocarbons have been suggested.  CSTR (continuously stirred tank reactor) and bubble column bioreactors (airlift bioreactor) containing immobilized cells were developed for the treatment of liquid phase pollutants.  However, the common shortcoming of these bioreactors was that pollutant was discharged to the environment by air injection for volatile organic compounds.

The purpose of the present investigation was to compare the biodegradation rates of a hydrocarbon mixture by two bioreactor systems, the bubble-column and the rotating biological contactors (RBCs), using a petroleum-degrading achlorophyllouis micro-alga, Prototheca zopfii.   The advantage of RBCs is their relative low energy consumption, simple operation and maintenance, and successive treatment of the influent contaminants.  As an alternative approach to treat hydrocarbons in bioreactors, the RBC appears to be a good choice because of these reasons.

MATERIALS AND METHODS

Microorganism and cultivation

The algal strain P. zopfii Krüger ATCC 30253 was used.  Hydrocarbon biodegradation using a mixture of n-tetradecane, n-pentadecane and n-hexadecane was tested according to previously published procedures [2].  The volumetric biodegradation rates observed in free cells were comparable to those reported in other systems of marine petroleum-degrading microorganisms, such as Bacillus sp. and Pseudomonas sp [2].  Biomass concentration measurements were carried out by counting the algal cells with a hemocytometer.

Cell immobilization

The most commonly used matrix for algae and cyanobacteria calcium alginate beads was not suitable for P. zopfii because of the strong hydrophobicity of the outermost surface of the algal cells and the nature of substrates, hydrocarbons [2].  Therefore, P. zopfii cells were immobilized naturally by physical entrapment within the open pore network of 8-mm-side polyurethane cubes (INOAC Co. Nagoya, Japan).  The average pore size of the particles used in the bubble-column type bioreactor was 0.83 mm.  Five pieces of polyurethane foam were added to flasks containing 10 ml basal medium that had been inoculated with algal cells and incubated for 7 days as in the previous study [2].  The total volume of the immersed particles was about 2.6 cm3.  Thereafter the microbes were successfully captured into the pores of the foam cubes and there were few suspended cells in the culture [3].

Bubble-column bioreactor

The bubble-column type bioreactor which was made of glass was 105 mm high and 36 mm in diameter and at the start of the experiment 10 pieces of foam were placed in the vessel with 50 ml medium.  The algal cells were inoculated into the reactor and culture was aerated by air through a sparger consisting of a straight tube of 1 mm diameter at the bottom of the reactor.  The reactor was immersed in a water bath thermostatted to ±0.1℃.  The culture was carried out at 25℃ with an aeration rate of 150 ml/min.  A mixture of tetradecane, pentadecane and hexadecane, each added at the initial concentration of 1% (v/v) [i.e. a total concentration of 3% (v/v)] was used as a model oil to be degraded.

Biodegradation in RBCs

The bioreactor was constructed from a reactor tank made of stainless steel, five rotating discs, and a motor.  The discs were made of polycarbonate and their diameter was 0.8 dm.  The total disc surface area available for microbial growth in the reactor was 5.0×10-2 m2.  The algal biomass was inoculated into the bioreactor containing 300 ml sterilized basal medium described previously [2] and the mixture of three types of alkanes was degraded as in the bubble-column bioreactor.  The submergence of the discs was about 30% and the rotational speed of the discs was 30 rpm.  The single stage RBC system was operated at 25℃ and at pH 7.0 in a batch mode.

 A cylindrical mesh drum filled with random packing of polyurethane cubes including biomass (biodrum) is used instead of a series of circular discs.  Three reactors can be used in series.

RESULTS AND DISCUSSION

Immobilized culture in a bubble-column bioreactor

After placing the foam cubes in the reactor, the cubes floated on the surface of the culture at the beginning of the experiments but remained in suspension after 15 min of cultivation when almost all algal cells were captured in the cubes.  Some leakage of growing cells was observed during the operation, however, significant cell leakage that had been observed in the wastewater treatment reactor for simultaneous removal of organic and nitrogenous substances [4], was not observed.

About 50% of hydrocarbons were utilized after 10 days.  The volumetric biodegradation rate of 30.3 (mg hydrocarbons / h per liter) was obtained from the data by compensating the amount of hydrocarbons evaporated (△).  From this observation, the additional facility such as bioscrubbers, trickling filters and biofilters should be combined with this type of bioreactor [5].

Biodegradation in the single stage RBC

Initial concentrations of both biomass and hydrocarbons were adjusted to be equivalent in the case of the bubble-column bioreactor.  The time courses of the amount of volatilized hydrocarbons in both systems can be seen from the data of control (without biomass) experiments, respectively.

About 65% of hydrocarbons were removed during 30 days of operation in the RBC (●).  The biomass on the disc reached about 1×104 cells/dm2 and was contained in a biofilm thickness of the RBC system would be influenced by many factors such as the types of organisms, speed of disc rotation, substrate concentration in the bulk liquid, roughness of the disc surface and temperature [6].  It appeared that the combination of P. zopfii and polycarbonate discs was a desirable choice since the alga had the strong hydrophobicity of the outermost surface of the cells [2]. 

Table 1 summarizes the comparison of removal rates by two reactors.  The net removal rates were evaluated by subtracting volatilized hydro-carbons from the removed hydrocarbons in the reactors including biomass. The volumetric biodegradation rate observed in the RBC system was reduced to about 60% of that in the bubble-column system. 

The validity of the proposed model was confirmed by comparing the simulated results with experimental data.  The kinetic and geometric values used in the simulation were listed in the previous study [7].

Modeling of RBCs in series

It is postulated that three RBC in series should be used for the treatment of the model hydrocarbons flowing parallel to the discs in continuous operation.  Primary effluent containing hydrocarbons enters section I7 of the first RBC and is degraded by both suspended and fixed biomass.  The treatment is repeated in the next stage.

where the value of Xs7,0 is 0 and the value of Ss7,0 is equal to the value of Sin.  The value of Sin was assumed to be equivalent to the initial hydrocarbon concentration treated in the present experimental study, 3% (v/v) or 2.8 kg m -3.  For the value of F, 1.25×10-6 m3 h-1, corresponding to a residence time of 10 days, was employed.  The initial thickness of the biofilm in each reactor was chosen to be equivalent to the experimentally determined value of 6.0μm in the single stage RBC for 30 days, and initial concentration of suspended biomass was assumed to be 0.

The concentrations of hydrocarbons in RBCs one to three increase in the early stage of cultivation and then concentrations decrease as the biomass concentration increases, respectively.  Eventually, the system reaches a steady state after about 60 days.  The inlet concentration of hydrocarbon mixture is increased suddenly from 22.8 to 45.6 kg m -3 at 150 days.

  An alternative RBC  can be identified as a hybrid reactor of the bubble-column type reactor using immobilized biomass and a standard RBC with biodisk.  The biodegradation rates observed in the RBC with biodrum were depend heavily on the physical properties of the polyurethane cubes.  The relationship between pore size and amount of algal cells immobilized within the cubes was examined for six kinds of polyurethane foam particles and select the most suitable one for the RBC system. 

CONCLUSION

Laboratory scale bubble-column and two types of RBC systems were utilized to treat a mixture of n-alkanes as a model hydrocarbons using P. zopfii in a batch operation.  The RBCs were potentially effective as an alternative approach to treatment using the bubble-column type reactor.   The performance of the typical RBC system with a series of discs was analyzed using a mathematical model which was based on mass balances for biomass and bio-degradable hydrocarbons in the reactor.